Bram Cohen talks about Bitcoin and Blockchain.
Bram Cohen is an American computer programmer, best known as the author of the P2P BitTorrent protocol, as well as the first file sharing program to use the protocol, also known as BitTorrent. He is also the co-founder of CodeCon and organizer of the San Francisco Bay Area P2P-hackers meeting, and was the co-author of Codeville.
“I think the actual amount of disruption that is happening because of bitcoin is really not much,” Bram Cohen recently said on the Steal This Show Podcast hosted by Jamie King. “If you look at bitcoin compared to [other] currencies, it’s still this flyspeck. People aren’t buying their coffee with bitcoin or anything like that. Some people have made a lot of money off it. There is the big story that a number of years ago buying $100 in bitcoin would be worth millions today.” That’s a story people get excited about, he acknowledges.
“In terms of what it’s actually doing, Bitcoin is still overhyped,” he opines. “It’s getting a lot more media attention than the actual impact it’s having so far.”
Mr. Cohen, who concerns himself with network protocol, says: “A lot of people get excited about bitcoin, because it’s gone up in value, so it’s the future and blah, blah, blah. Most of these people don’t have any understanding about what Bitcoin actually is.”
But eventually people Mr. Cohen trusts pointed out to the seasoned distributed developer, ‘No, there is actually an interesting bit of cryptographic stuff’.
He saw the light: “And, yeah, there is a really interesting thing in there.”
But, people make outrageous claims, Mr Cohen points out. “‘Oh, it has infinite scaling at no cost and instant transactions and its like, ‘No, No, you are not helping. You’re spouting a bunch of bullshit and those statements are just wrong.”
He acknowledges the problems bitcoin solves.
“Bitcoin is a distributed database,” he elucidates. “So, the history here is there is this problem with doing a K agreement in databases. Basically, you can’t have a distributed database. You can prove that you can’t really have a distributed database. But, it turns out there is a loophole in this and you can have a distributed database, if you make it so your database can only handle a few transactions per second using just about the simplest semantics a database can support, a replication strategy that is everything goes everywhere, it takes half hour for anything to go through, and you have warehouses doing nothing whatsoever but burning electricity. Then you can have a distributed databases. And that is Bitcoin.”
Mr. Cohen’s passion is the details of protocols. “When you have a situation where everything is broken, and horrible, there is lots of stuff to work on. There is all sorts of stuff you can do here to try and improve the situation.”
Bitcoin could use some improving, he believes. The original crypto-currency is inefficient, because Bitcoin’s mining power is based on processor time. Mr. Cohen believes basing the value of bitcoin mining on storage is a more elegant solution.
By using the storage based solution, Mr. Cohen believes he will address multiple Bitcoin flaws. In a storage based system, there’s a lot less centralization in mining.
“So there’s a lot less concern about having a 51% attack,” Mr. Cohen says. “Sometimes people have this misapprehension that Bitcoin is a democracy. No, Bitcoin is not a democracy; it’s called a 51% attack for a reason.”
He believes in their pursuit of private, permissioned ledgers, banks will come up empty-handed. “Banks are looking at it and are kind of freaked out,” he says. “A number of banks are like, ‘We need blockchain, because it is the future. We try and explain to them, ‘Well, this doesn’t make a lot of sense for what you’re talking about.’”
He is skeptical about the novelty of permissioned blockchains, which use secure hashes in the backend.
“What they wind up doing is a term of art, a ‘private blockchain’, database using secured hashes in the backend,” Mr. Cohen notes. “Vendors are selling such things, and I think doing quite good business off that. And they say we have blockchain now. [It] has nothing to do with blockchain. Yeah, it’s a huge improvement over what you were doing before but you should have done that twenty years ago,” Bram believes. “Secure hash based auto trails are not a new thing.”
Check out the full interview here.